- Email: [email protected]
Chemical stimulation for enhancing coal seam permeability: Laboratory study into permeability variation and coal structure examination
International Journal of Coal Geology 219 (2020) 103375
Contents lists available at ScienceDirect
International Journal of Coal Geology journal homepage: www.elsevier.com/locate/coal
Chemical stimulation for enhancing coal seam permeability: Laboratory study into permeability variation and coal structure examination Zhenhua Jinga,b, Reydick D. Balucanb, Jim R. Underschultzc, Songqi Pand, Karen M. Steelb,
T
⁎
a
School of Earth and Space Sciences, Peking University, Beijing 100871, China School of Chemical Engineering, The University of Queensland, St Lucia 4072, Australia c Centre for Natural Gas, The University of Queensland, St Lucia 4072, Australia d Research Institute of Petroleum Exploration and Development, PetroChina, Beijing, 100083, China b
A R T I C LE I N FO
A B S T R A C T
Keywords: Coal seam gas Permeability Acid stimulation HCl Oxidant stimulation NaClO
Many potential coalbeds have sub-economic permeability for coal seam gas (CSG) extraction even though all the other key characteristics (gas content, thickness etc.) can meet the requirement for successful production. To enhance coal seam permeability various CSG stimulation techniques, including hydraulic fracturing, cavity well completions and horizontal wells, have been used, but none is ubiquitously successful. The potential of chemical stimulation including both acid (1% HCl) and oxidant (1% NaClO) stimulation are examined in this paper as an alternative method to enhance coal seam permeability. Acid stimulation targets the removal of certain minerals in coal cleats, while the oxidant dissolves the coal matrix, thereby etching pre-existing coal cleats and perhaps forming new ones. The permeability variation during stimulations are measured with core flooding tests and the corresponding coal structural changes are examined using X-ray microcomputed tomography (μCT) technique. The chemical reaction mechanisms are confirmed by elemental analysis via inductively coupled plasma optical emission spectroscopy (ICP-OES) and dissolved organic carbon (DOC) in the effluent. Acid stimulation was performed horizontally with cube samples cut from coal cores (Bowen Basin, Australia) and it exhibited a positive effect on coal permeability, which is proposed to result from demineralisation as shown by before and after CT scanning observations together with high calcium (Ca) content in the effluent. NaClO oxidation was able to etch pre-existing coal cleat surfaces, widen cleat apertures and generate new horizontal fractures and/or void space. These changes were caused primarily by chemical attack, confirmed by the high DOC concentration in the effluent. Oxidant stimulation caused a decrease in vertical permeability and an increase in horizontal permeability for some cores. The decrease in permeability is proposed to be due to a combination of increased void space weakening the coal associated with movement of the coal into newly created space against confining pressure. Minerals in the cleats appear to play a role keeping void space open and preventing collapse under confining pressure. Furthermore, the results for these samples indicate that NaClO oxidation appears to be lithotype independent and preferentially attacks coal sections that have more initial pores or fractures.
1. Introduction Commercial coal seam gas (CSG) extraction has been well established in a number of countries throughout the world, including the USA, Australia, China, India and Canada (Underschultz, 2016). The permeability of coal is a key controlling factor for gas migration in coal reservoirs and is often a critical technical barrier preventing economically viable gas production (Kumar et al., 2011; Moore, 2012). Many low permeability CSG reservoirs require stimulation to allow economic development, even if the seams contain high gas contents (Colmenares
⁎
and Zoback, 2007; Gamson et al., 1996; Laubach et al., 1998; Liu and Harpalani, 2013; Moore, 2012). Although various CSG stimulation techniques, including hydraulic fracturing, cavity well completions and horizontal wells (Colmenares and Zoback, 2007; Palmer, 2010; Palmer et al., 1993), have been used commercially, their success in some wells remain unsuccessful (Johnson et al., 2010; Palmer et al., 1993; Zou et al., 2014). Alternative stimulation options are desirable for increasing gas extraction from coal seams. Chemical stimulation has recently been studied, including both acid and oxidant stimulation (Balucan et al., 2016; Jing et al., 2018a; Turner
Corresponding author. E-mail address: [email protected] (K.M. Steel).
https://doi.org/10.1016/j.coal.2019.103375 Received 16 September 2019; Received in revised form 18 December 2019; Accepted 18 December 2019 Available online 20 December 2019 0166-5162/ © 2019 Published by Elsevier B.V.
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
enabled us to identify the coal structure modification and facilitate understanding of different stimulation effects and mechanisms.
and Steel, 2016). The former targets the removal of minerals from the cleats, while the latter is directed at dissolving the coal matrix to etch the existing cleat surfaces and increase cleat apertures and/or generate new fractures (Jing, 2019). In addition, weakening the coal matrix by oxidant reactions may also prove applicable as a pre-treatment to enhance hydraulic fracturing performance. For low permeability coal seams with highly mineralised fractures, dissolution of mineral occlusions will likely increase fluid flow capacities (Balucan et al., 2016; Turner and Steel, 2016). Vasyuchkov (1985)reported aqueous HCl (2, 4 and 6%) solutions could enhance the coal permeability dramatically from less than 0.1 mD to over 15.1 mD. Turner and Steel (2016) conducted core flooding tests with coal samples containing various cleat and mineralogical properties. The resulting permeability enhancement was “mineral type and rate of occurence” dependent that the largest increases in permeability were achieved where cleats were rich in carbonates, however, the presence of clays placed limitations on the extent of enhancement. Balucan et al. (2018) investigated the acid-induced physical and chemical changes in core samples to gain improved understanding of permeability enhancement in the vertical flow direction. There was clear evidence for cleat opening via mineral dissolution as well as mineral mobilizationaccumulation. Based on the X-ray microcomputed tomography (μCT) image analyses, they conducted flow simulation experiments and reported that acidizing could enhance permeability not only in the vertical direction but also in the lateral or horizontal direction (Balucan et al., 2018). Similarly, Ramandi et al. (2018) applied the μCT and numerical simulation techniques to investigate the effect of dissolution of syngenetic and epigenetic minerals on coal permeability and reported that coal permeability increase depends significantly on epigenetic mineral dissolution, and secondarily on syngenetic mineral dissolution. Although the above studies have examined the effect of acidization on coal permeability experimentally and numerically, uncertainty still remains. For example, few of their results relate to the horizontal permeability. Additionally, in the core flooding tests conducted by Turner and Steel (2016), the HCl was injected in the same direction with brine. However, to mimic how stimulation would be conducted downhole, forward flow to characterise initial permeability followed by reverse flow with acid to simulate the downhole treatment and finally forward flow again to characterise the final permeability is required. In addition to acid stimulation, we have recently reported on oxidant stimulation of coal seams (Jing et al., 2018a,b) where NaClO was determined to be one of the most effective oxidants, given its high reactivity and soluble reaction products (Liu et al., 2013a,b; Yao et al., 2010). The action of NaClO in artificial cleats on natural coal surfaces revealed increases in the cleat aperture (Jing et al., 2018b). Furthermore, NaClO oxidation appeared to increase the coal porosity by enlarging the pore sizes, which was measured by mercury intrusion porosimetry and observed by Scanning Electron Microscopy (SEM) imaging (Jing et al., 2018b). The matrix pore size enlargement and the cleat aperture dilation are proposed to be capable to increase the cleat system connectivity, thus increasing coal seam permeability. In this paper, coal core samples containing fully developed cleat systems were used to explore the effect of acidisation and oxidation on coal permeability under confining pressure. The permeability was measured using the core flooding rig which enabled the forward flow followed by a reverse flow and a final forward flow again to mimic how stimulation would be conducted downhole. The cubic samples cut from the cores allowed us to measure horizontal permeability variation. In terms of the oxidant stimulation, previously we have reported on a screening procedure which identified NaClO (Jing et al., 2018a) as the most promising oxidant and examined its effect on coal structures including pores and cleats (Jing et al., 2018b), Here we report on the coal permeability variation after NaClO treatment, both vertically and horizontally, to investigate the efficacy of NaClO to increase coal permeability. Furthermore, CT scanning of samples before and after treatment
2. Methodology 2.1. Coal samples The chemical stimulation in this study included acid stimulation with 1% HCl and oxidant stimulation with 1% NaClO. Stimulation was examined perpendicular and parallel to the coal bedding plane. The vertical permeability was studied on coal cores (76 mm diameter) collected from Bowen Basin (labelled BV), while the horizontal permeability was measured with cube samples (40 × 40 × 40 mm) which were cut from cores from the same basin but different boreholes (labelled BH). Core sample information including sample size and mineral information has been previously reported (Turner and Steel, 2016). The two coals shared similar vitrinite reflectance that was 0.86% for BV coal and 0.84% for BH coal. Their mineral compositions were slightly different in that BV coal was primarily comprised of kaolinite while BH coal was mainly composed of carbonates (calcite, siderite) and barite and secondarily of kaolinite. In addition, BV coal and BH coal both contain dull and bright coal bands as evidenced by visual observation and CT scanning. The acid stimulation for coal cores BV1 and BV2 had been conducted with 1% HCl previously where they were labelled “A1” and “A2”, respectively (Turner and Steel, 2016). In this work, these cores have been treated with NaClO. Cube samples (BH samples) were taken from the same borehole as the samples labelled “C” in previous research (Turner and Steel, 2016). In this work, these cores were treated with HCl and NaClO. 2.2. Core flooding test Steady state core flood permeability testing was performed using the core flooding rig system(Fig. 1) and it was updated from the rig previously reported by Turner and Steel (2016). The system mainly includes a Hassler core holder; one high-pressure Quizix QX6000 dual cylinder pump for fluid injection; one Redox gas/oil tank for confining pressure control; Gems 3200 series pressure transducers (0–100 bar) to measure both pore pressure drop and confining pressure; and a domeloaded back pressure regulator to maintain fluid pressure drop across the sample. Wetted parts were predominantly Hastelloy to provide superior corrosion resistance to brine, dilute acids and oxidants. The components in the system have been reported previously (Balucan et al., 2016; Turner and Steel, 2016). Instead of flooding brine and the stimulants in the same direction, the case in previous research by Turner and Steel (2016), this system allowed the stimulant to flow in a reverse direction to mimic the stimulation procedure in-situ. Brine and acid were injected through the sample by the pump, while compressed air was used to push oxidant into the coal samples in order to protect the pump from oxidation. For each run, the oxidant volume flowing through the coal was deliberately monitored to be less than the oxidant cylinder volume to prevent gas from entering the coal samples. Prior to the flooding test, the coal samples were X-ray CT scanned to characterise the coal structures, particularly the fracture systems. The flooding tests start with the coal samples saturated in brine (4% KCl) to enable the coal to be at equilibrium with in-situ conditions and consolidated via repeated loading/unloading cycles. Once the saturation was established, a forward brine injection was conducted until the permeability measurement stabilised, which established the initial permeability baseline. Afterwards, the stimulant (either 1% HCl or 1% NaClO) was injected in the opposite “reverse” direction until the permeability stabilised under the same circumstances, i.e. same confining pressure (Pc), inlet pressure (P1), outlet pressure (P2) and effective pressure (Pe). This was followed by a second forward flow with brine to compare the permeability change before and after stimulation. 2
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
Fig. 1. Schematic of core flooding system. Table 1 Pressure system of the coal flooding test for each sample. Sample
BV1 BV2 BH1 BH2 BH3
Stimulant
1% 1% 1% 1% 1% 1% 1%
NaClO & 4% KCl NaClO & 4% KCl HCl & 4% KCl NaClO & 4% KCl HCl & 4% KCl NaClO & 4% KCl NaClO & 4% KCl
Flooding direction
Pressure system
Vertical Vertical Horizontal Horizontal Horizontal Horizontal Horizontal
P1 (bar)
P2 (bar)
Pc (bar)
Pe (bar)
10 10 15 10 15 10 10
4 4 4 4 4 4 4
25 25 25 25 25 25 50
18 18 15.5 18 15.5 18 43
Image segmentation into three coal components, namely unfilled fracture (voids or fracture porosity), coal matrix and minerals, was achieved via greyscale unit (GU) thresholding. Components were allocated specific greyscale ranges following detailed surveys of the greyscale values of visually discernible structures. Fracture or mineral fractions (Xf) of individual slices were estimated using Eq. (1), where ∑PXGUrange is the sum of pixels within the greyscale range (lower to upper limit) of the component and ∑PX0−255 is the total pixels for the entire slice. Following estimates validation, the core profiles were compared to identify the structural changes of coal structures.
Subsequently, the sample was X-ray CT scanned again to investigate the coal structural change after the flooding test. The pressure details about the core flooding tests are summarised in Table 1. Liquid samples were regularly extracted from the core holder outlet and analysed for element concentrations via inductively coupled plasma optical emission spectroscopy (ICP-OES) and for dissolved organic carbon (DOC). The procedure for ICP-OES and DOC analysis has been previously described (Jing et al., 2018a; Turner and Steel, 2016). 2.3. X-ray computed tomography (CT) scanning An Inveon Multimodality PET/CT scanner (Seimens) at the Centre for Advanced Imaging (CAI), The University of Queensland, was used (voltage = 80 kV, beam current = 0.5 mA) to image the coal sample before and after stimulation. Scanned images of the core were reconstructed, visualised and normalised using the Inveon™ Research Workplace software (Seimens IRW v4.2). The 3D reconstructions and visualisation enabled qualitative surveys. These surveys were fundamental to understanding core structure that is essential for validating the generated core profiles. Greyscale image normalisation, attained by keeping the contrast parameters constant, allowed for comparison of the images before and after stimulation. These axial greyscale images (display resolution = 44 μm) were then used for mapping the mineral and fracture profiles of the core before and after stimulation. Core profiling involves image stack generation, slice histogram extraction, image segmentation and fractionation. The image stacks each containing ~1600 greyscale axial slices (images) were manually registered, stabilised and cropped prior to extracting the slice histograms using plugins and macros embedded in ImageJ software.
Xf =
∑ PXGUrange ∑ PX0 − 255
(1)
3. Results 3.1. Effect of oxidant stimulation on vertical permeability Two coal cores, BV1 and BV2, which have been previously treated with 1% HCl were used to examine the effect of oxidation on vertical permeability. 3.1.1. Coal core sample BV1 Coal BV1 was initially flooded perpendicular to the bedding plane by 4% KCl (forward flow) until the permeability reached a plateau and then 1% NaClO diluted with 4% KCl was injected in the opposite direction (reverse flow) to stimulate the coal until the permeability stabilised. Afterwards, a second forward flow with 4% KCl was conducted to examine the permeability change by comparing with the initial 3
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
same site of the upper coal core before and after oxidation, while Fig. 5c and g are from the same position of the lower section. The apparent change could be observed by comparing Fig. 5b and f. After oxidation, the void space in Fig. 5f accounted for 30% of the whole image and a horizontal fracture existed on the left of the image. In the lower section, by contrast, the structure primarily kept intact without discernible change (Fig. 5c and g). The different effects of oxidation on upper and lower sections can be further demonstrated by the VF distribution (Fig. 5h), where the VF in the upper section after oxidation (green line) is considerately higher than the initial VF (Blue line), while in the lower section the two lines overlapped. 3.2. Effect of acid and oxidant stimulation on horizontal permeability
Fig. 2. Permeability variation of BV1 in 1% NaClO stimulation. P1 = 10 bar, P2 = 4 bar, Pc = 25 bar.
Cube samples enable the measurement of horizontal permeability parallel to the coal bedding plane, which reflects the in-situ permeability direction parallel to bedding and perpendicular to the borehole which is the desirable direction to examine the permeability for a wellbore stimulation effect. Few researchers have reported the effect of acid stimulation on horizontal permeability and even less the effect of oxidation. In this study three cube samples were used and two of them (BH1 and BH2) experienced acid stimulation first and oxidant stimulation afterwards, while the other one (BH3) underwent oxidant stimulation only.
forward flow. The permeability variation during this procedure was illustrated in Fig. 2 for coal BV1. The initial stable permeability with 4% KCl was 0.59 mD and gradually dropped to 0.02 mD during the injection of 230 mL NaClO. Afterwards, the forward permeability post oxidation stabilised at 0.14 mD. To understand the permeability decline, the coal core structural change induced by oxidation was examined based on CT scanning data as shown in Fig. 3 where the core flooding was directed from top to bottom. Compared with the initial coal core 3D image (Fig. 3a), NaClO oxidation induced a group of fractures occurring at the middle of the coal core (Fig. 3e). The cross-section CT images could further illustrate the fracture pattern from a top to bottom view. On Fig. 3f, the fractures or void spaces, which did not exist pre oxidation (Fig. 3b), extended horizontally through the coal core. The void space fraction (VF) after oxidation increased significantly, particularly in the middle location of the coal (Fig. 3g) where it was initially more porous than other parts in the core (Fig. 3c). In the same section, the 3D image of the oxidised coal core revealed a roughly horizontal connected layer as shown by the blue layer in Fig. 3h.
3.2.1. Cube sample BH1 Acid stimulation with 1% HCl and 4% KCl was firstly conducted for coal BH1. The corresponding permeability variation and the element concentrations in the effluent are shown in Fig. 6. The permeability of initial forward flow was 0.033 mD. When 1% HCl was flooded in the reverse direction, its permeability increased to 1.2 mD but then gradually declined to 0.38 mD with a total 1550 mL 1% HCl flowing through. The following forward flow started with a permeability of 0.8 mD and eventually reached 0.35 mD, which was more than 10 times higher than the initial forward permeability (0.033 mD). The increase of permeability during HCl flooding was accompanied by a drastic elevation in the calcium (Ca) concentration and a minor iron (Fe) increase as shown in Fig. 6, indicating the significant dissolution of calcite (CaCO3) and a slight dissolution of ankerite (Ca(Fe, Mg, Mn)(CO3)2). The concentration of aluminium (Al) and barium (Ba) were quite low. After 1% HCl flooding, the coal was subjected to oxidant stimulation with 1% NaClO and Fig. 7 shows the permeability and DOC concentration variation during this procedure. The initial forward permeability was 0.038 mD. Note that the initial forward permeability in 4% KCl (F1) was not equal to the final forward flow (F2) in the acid stimulation because they did not share the same pressure conditions (Table 1). The reverse permeability during oxidant injection started at 0.118 mD and generally dropped to 0.04 mD after 60 mL 1% NaClO flowing through. The forward flow post oxidation showed a horizontal permeability of 0.003 mD, revealing the damage to coal flow capacity caused by oxidation. In Fig. 7, the DOC concentration in the initial forward flow with 4% KCl was 14 mg/L, and drastically increased with the injection of NaClO to 350 mg/L. Afterwards, it dropped to 26.62 mg/L in the second brine forward flow. The high DOC concentration in the effluent during NaClO injection demonstrated the chemical attack happening to the coal matrix. Fig. 8 shows the coal structural change associated with both acid and oxidant stimulation based on CT scanning images. The CT image slides were collected and shown in two directions: Y direction, which is the horizontal flooding direction; Z direction, which is the top to bottom vertical direction. In the Y direction, the CT images from the same site of the coal pre stimulation, post acid stimulation and post oxidant stimulation are shown in Fig. 8b, c and d, respectively. The sites
3.1.2. Coal core sample BV2 The second core sample (BV2) was pre-treated with 30% H2O2 because H2O2 has been reported to be able to enhance the oxidative effect of NaClO for coals (Mayo, 1975). Coal BV2 experienced two stages of oxidation, 30% H2O2 and then 1% NaClO. The corresponding permeability variation with each stage is shown in Fig. 4. The permeability of initial forward flow (F1) with brine (4% KCl) was 2.01 mD, followed by the reverse (R1) with 30% H2O2 (total 40 mL), where the permeability dropped to 0.08 mD. A second forward flow (F2) was conducted with a measured permeability of 0.59 mD, showing the detrimental effect of 30% H2O2 on permeability. Afterwards, 50 mL of 1% NaClO was injected into the same sample, and the permeability was recorded to steadily decline from 0.25 mD to 0.05 mD. Subsequently a final forward flow (F3) showed the permeability stabilised at 0.17 mD, much lower than the initial 2.01 mD. Dissolved organic carbon (DOC) dissolved in the effluent solutions showed the concentration in brine was low at 23.5 mg/L (Fig. 4). During the H2O2 flooding, DOC increased slightly to 31.8 mg/L. DOC concentration in the effluent increased with injection of NaClO. It then dropped during the final brine flooding. The structural variation of sample BV2 before and after treatment with 30% H2O2 and 1% NaClO are shown in Fig. 5. The coal core was divided into upper and lower sections (Fig. 5a). In the upper section, more void spaces existed initially, as shown in Fig. 5d illustrating the void space fraction (VF) distribution along the coal core axis, while the lower section was physically tighter. After oxidation treatment, massive coal structural change occurred in the upper section, rather than its lower counterpart (Fig. 5e). This could be clearly illustrated by the cross-section images. Fig. 5b and f are the cross-section CT images at the 4
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
Fig. 3. BV1 core structure analysis based on X-ray μCT scanning. a. Vertical cross-section across core length before oxidation; b. horizontal slice through image shown in (a); c. void space fraction along core length before oxidation; d. 3D image of mineral and void space in before oxidation; e. Vertical cross-section across core length after oxidation; f. horizontal slice through image after oxidation at the same location as that shown in (b); g. void space fraction along core length after oxidation; h. 3D image of mineral and void space in the core after oxidation;
(Fig. 8b) vanished after acid stimulation (Fig. 8c). Similarly, the contents of minerals in the dull bands could be observed to decrease albeit less obviously. Then after oxidant flooding, the site shows as a vacant fracture (Fig. 8d). The variation of fracture features explicitly indicates the demineralisation effect during acid stimulation and the etching effect of oxidant on fractures. These fracture variations were confirmed by viewing CT images in the Z direction as shown in Fig. 8e (pre-acid), Fig. 8f (post-acid) and Fig. 8g (post-oxidant). The images are a cross section selected in the bright coal band (red rectangle in Fig. 8a), which was initially highly mineralised (Fig. 8e). The minerals filling in the cleats highlighted by the red ellipse were removed after acid stimulation and a void cleat appeared at the same section after oxidation. Moreover, after oxidation, the vacant cleats formed a connected fracture system across the sample. Fig. 8h, i, g represent the 3D image of both void space and minerals in the coal in the Z direction (viewing from top to bottom in Fig. 8a), before stimulation, post acid and post oxidant stimulation, respectively. The minerals were extensively distributed in the whole sample before acid stimulation, but after acid flooding, the amount of minerals decreased, especially in the middle of the sample. After oxidation, an
Fig. 4. Permeability and DOC concentration variation at different oxidation stages for coal BV2. P1 = 10 bar, P2 = 4 bar, Pc = 25 bar.
selected by the red dash rectangle in these figures exemplify a typical change of the coal structure undergoing the stimulations. The minerals initially filled in the fractures in the bright coal band before stimulation 5
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
Fig. 5. BV2 sample coal structure analysis based on core CT scanning. a. 3D image before oxidation; b. horizontal cross-sectional slice in upper section before oxidation; c. horizontal cross-sectional slice in lower section before oxidation; d. void space fraction before oxidation; e. 3D image after oxidation; f. cross-sectional slice in same spot as (b) after oxidation; g. cross-sectional slice in same spot as (c) after oxidation; h. comparison of void space and mineral fraction before and after oxidation. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)
may be the vacant cleat not filled by minerals. The initial mineral distribution was reflected in Fig. 9b (blue line) with two humps showing the mineral fractions, while the initial void space was shown in Fig. 9c (blue line). After acid stimulation, the drop of mineral fraction is clearly shown in Fig. 9b where the original two humps disappeared and the mineral fraction stayed at a similar low level along the whole coal sample. This means most of the minerals were removed by 1% HCl. By comparison, the void space created is also shown as two humps at the locations where minerals existed initially (red line Fig. 9c). Afterwards, the oxidation further enlarged the void space regardless of the coal type, as shown in Fig. 9c. NaClO appeared to barely influence the minerals. Overall, the results clearly quantified the demineralisation of acid stimulation and the enlargement of void space after oxidation. However, the increase of void space after oxidation detected by CT imaging could not explain the bulk decline of permeability.
Fig. 6. Permeability and element concentration change during 1% HCl stimulation for coal BH1. Ca: Calcium; Fe: Iron; Ba: Barium; Al: Aluminium; P: Phosphorus. P1 = 15 bar, P2 = 4 bar, Pc = 25 bar.
3.2.2. Cube sample BH2 Sample BH2 shares a similar lithotype with BH1 that contains both dull coal and bright coal layers. The same stimulation procedures were conducted to BH2, i.e. acid stimulation first followed by oxidant stimulation. Fig. 10 illustrates the permeability variation and the element concentration in the flooding effluent for coal BH2 during 1% HCl stimulation. The initial forward flow had a stable permeability of 0.35 mD, which generally increased to 1.21 mD after 1700 mL 1% HCl flowed through. Afterwards, the post-stimulation forward flow stabilised at 0.85 mD. The permeability variation shows the 1% HCl stimulation could elevate the permeability 2.42 times for coal BH2. Upon injection of HCl, there was a spike in the concentration of Ca in the effluent, from 65 ppm in the initial brine flow to 500 ppm in the 1% HCl flooding. Then it decreased dramatically with increasing HCl injection presumably due to complete removal of accessible calcite (CaCO3). The concentrations of P and Fe were significant, indicating dissolution of apatite (Ca5(PO4)3(F,Cl,OH)) and ankerite (Ca(Fe, Mg, Mn)(CO3)2). Concentration of Ba mainly derived from barite (BaSO4) was much higher in the initial brine flooding for coal BH2 than that for coal BH1, indicating a high barite content existing in the coal structures. Barite was observed by SEM in the coal cleats after acid stimulation, as shown in Fig. 11. Dissolution of apatite could produce weak HF which could then
Fig. 7. Permeability variation during 1% NaClO stimulation for coal BH1. P1 = 10 bar, P2 = 4 bar, Pc = 25 bar.
apparent void fracture could be observed across the whole sample in the XY plane (horizontal plane). Additionally, the void space fraction (VF) and mineral fraction (MF) change along the coal sample from top to bottom (z direction) further demonstrating the demineralisation effect of acid stimulation and the etching effect of oxidant stimulation as shown in Fig. 9. Fig. 9a illustrates two distinct regions of which the upper part is dull coal and the lower part appeared to be bright coal filled with cleats. Some void space (shown as blue colour in Fig. 9a) could be noted in the bright coal and it 6
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
Fig. 8. BH1 coal structure change after acid and oxidant stimulation. a. A diagram of coal BH1 showing the stimulation direction is along Y; b, c and d show the same image viewed in the stimulation direction before and after acid and oxidant stimulation. e, f and g show images of the cross-section shown in a; h, i and j show 3D images for coal BH1 before and after stimulations from a view angle in z direction.
After acid stimulation, a forward flow was conducted with an inlet pressure reduced to 10 bar to build the permeability baseline for oxidant stimulation, which was measured as 0.14 mD. Then the 1% NaClO flooding followed, where the reverse direction permeability reached quite high at the beginning (7.5 mD) and declined to 2.3 mD with a
dissolve kaolinite (Al2Si2O5(OH)4)). Significant levels of Al were observed at the end of acid flooding. CT scanning images failed to explain the coal BH2 structural change due to the X-ray beam hardening caused by extensive mineral content, particular the high concentration of barite. 7
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
Fig. 9. Void space fraction and mineral fraction for coal BH1. a. 3D image for coal BH1 showing mineral and void space before oxidation and the flooding direction was along Y; b. Mineral fraction before and after acid and oxidant stimulation; c. Void space fraction before and after acid and oxidant stimulation. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)
Fig. 10. Permeability variation in 1% HCl stimulation for coal H2. Ca: Calcium; Fe: Iron; Ba: Barium; Al: Aluminium; P: Phosphorus.P1 = 15 bar, P2 = 4 bar, Pc = 25 bar.
Fig. 12. Permeability variation in 1% NaClO stimulation for coal BH2. P1 = 10 bar, P2 = 4 bar, Pc = 25 bar.
Fig. 11. SEM image of Barite in the cleat of coal BH2 after acid stimulation. a. the Barite on the inlet surface of BH2; b. Barite in the cleat of the inlet of BH2. 8
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
coal to keep its minerals intact. Fig. 13 shows the horizontal permeability variation during the single oxidant stimulation. The initial permeability was 1.95 mD, which dropped to 0.62 mD at the beginning of oxidation, and after 1 hour flowing, the permeability shot up to 8.2 mD and fluctuated around this value. Then the forward permeability of brine was 6.84 mD, 4.38 times higher than the initial (1.95 mD). CT images pre/post oxidation for coal BH3 revealed the coal structural changes caused by oxidation as shown in Fig. 14. Combining the hand sample observation and the 3D image (Fig. 14a), it could be confirmed that there existed multiple layers of bright coal and one dull coal layer with an initial fracture across shown as a blue layer. After oxidation (Fig. 14c), void space proportion could be viewed increasing in the middle of the sample. Fig. 14b and d show the cross-section CT images of the initial fracture in dull coal before and after oxidation, respectively. The comparison of these images could identify two fracture feature changes that could benefit the permeability. The first was the increase of fracture aperture. Before oxidation, the fracture apertures were measured as 134 μm and 126 μm for spot 1 and spot 2, respectively, and they were enlarged to 174 μm and 228 μm after oxidation. The second was the extension of the existing fracture, which was illustrated in Fig. 14d selected by the red rectangle, where a new fracture was generated and connected to the initial fracture.
Fig. 13. Permeability variation during 1% NaClO oxidation for coal BH3. P1 = 10 bar, P2 = 4 bar, Pc = 50 bar.
total flooding volume of 1300 mL. The post-stimulation forward flow started at 3.5 mD and ended with 0.71 mD, which was 5 times higher than the pre-oxidation permeability as shown in Fig. 12. DOC concentration in the initial forward flow was 31 mg/L and shot up to 108 mg/L at the beginning of NaClO injection followed by a gradual decrease. The final brine flow was 19 mg/L. This DOC trend for coal BH2 was different to that for coal BV1 and BH1, where the DOC concentration gradually increased during the NaClO injection. The difference is proposed to be caused by the permeability level during the NaClO flooding, which was significantly higher here than that for coal BV1 and BH1. The higher permeability allows the dissolved organic carbon to flow out immediately, while the lower permeability hampers the flow and delays the detection of dissolved coals. Furthermore, the average DOC concentration for coal BH2 was lower than those for coal BV1 and BH1. This is caused by the significant volume of NaClO flowing through coal BH2, which reduced the average DOC concentration in unit volume (10 mL here).
4. Discussion The horizontal permeability of the two studied cube samples (BH1 and BH2) both increased after acid stimulation. The CT image comparison for coal BH1 before and after oxidation (Fig. 8) showed that minerals were removed in some parts of the cleats after 1% HCl flooding and the mineral fraction (MF) calculation based on CT images confirmed the demineralizing effect. New void space occurred at the locations where initially minerals existed. These generated void spaces may stay open during the flooding test due to the hydraulic pressure and provide pathways for fluid thus elevating the permeability during acid stimulation. The ICP-OES results for the two studied samples confirmed the demineralisation caused by 1% HCl flooding. For both coals, Ca was the main detected element in the effluent, indicating the dissolution of Ca containing minerals including calcite, aparite and ankerite. These results of an increase in horizontal permeability agree with those found
3.2.3. Cube sample BH3 Both of the two cubic samples described above underwent acid stimulation followed by oxidant stimulation. Oxidation showed a detrimental effect on BH1 permeability, but a positive effect on BH2 which contained a significantly higher mineral content, especially with respect to Barite. This may indicate that minerals play a role in oxidant stimulation. Therefore, for coal BH3, only oxidant was flowed through the
Fig. 14. Fracture feature change in coal BH3 after 1% NaClO oxidation. a.3D image for coal BH3 before oxidation where blue shows a large fracture across a dull coal band; b. Cross-section of the large fracture and where aperture was measured at two spots; c. 3D image for coal BH3 after oxidation; d. Cross-section of the large fracture after oxidation. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) 9
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
examine this conflict, the void space fractions for each coal before and after oxidation were analysed and we found that the NaClO attacks in the locations where the void space is highest. The extent of attack appears to be dependent on the surface area. The NaClO oxidation therefore appears to be lithotype independent and the NaClO attacks any coal type as long as it can come to contact with surface area. For example, in coal BH3, where a fracture existed in the dull coal band, coal oxidation not only widened the fracture aperture, but also produced a new fracture.
for vertical permeability for Ca-rich cores (Turner and Steel, 2016), where increases in permeability were observed. After oxidation, the vertical permeability decreased for both coal cores, while the horizontal permeability decreased for BH1, but increased for BH2 and BH3. Analysis of CT scanning images showed three main changes on coal structures after NaClO oxidation. The first is the enlargement of the void space at locations that were initially more porous. The second is the generation of new fractures close to the horizontal bedding direction or the extension of existing fractures. The third is the general enlargement of fracture apertures. These changes could suggest an increase in permeability. For the coal cores BV1 and BV2, a substantial decrease in vertical permeability was observed. It is proposed that the bulk vertical permeability is determined by the section having the lowest permeability. Furthermore, it is proposed that oxidant stimulation weakened/softened the cores. Against a confining pressure, it is proposed that the softened coal collapsed in on the newly formed void spaces, leading to the drop in permeability. For coal cores BH1 and BH2 a decrease and increase in permeability was observed. The mineral difference between BH1 and BH2 may have led to the differences in permeability after stimulation. BH1 and BH2 both experienced 1% HCl flooding, which dissolved a large proportion of minerals. During oxidant flooding, although coal fractures in BH1 were etched and enlarged, the confining pressure may have caused the closure of existing fractures with the decline in fluid pressure (Balucan et al., 2016; Kumar et al., 2011) and even result in the fractures collapsing. It is worthwhile to note that confining pressure is applied from all sides in the set-up (Fig. 1), which may not adequately represent the stress conditions downhole. Therefore, it is uncertain whether collapse in enlarged fractures would occur. For BH2 a significant amount of carbonate minerals were also dissolved and a considerable amount of barite remained. This barite may have helped to maintain the coal structure integrity (i.e. prop open fractures against the confining pressure conditions). The large amount of barite could also support the cleats during the oxidant flooding (Fig. 11). To examine this hypothesis, coal BH3 was directly flooded by oxidant without a pre-treatment with acid to keep the minerals intact. The permeability for BH3 increased from 1.95 mD to 6.84 mD (Fig. 13). The void space proportion in the bright layers was confirmed to be increased (Fig. 14) and the remaining minerals may have protected the cleats from collapsing. Besides the minerals, other factors could affect the permeability during oxidation including specific coal fracture structures, the oxidant volume, the contacting time and pressure. For coal BH1, although the flooding direction was parallel to the face cleats, the face cleats terminated in the middle of the coal and connected with another face cleat by a butt cleat. The connection points with a narrow aperture might act as a restriction point. Coal fines could accumulate here and halt the permeability enhancement. For coal BH3, the face cleats went through the whole sample, making it easier for fluid to flow through without fines jamming. In addition, coal BH3 contained a large fracture which could allow the fluid to flow quickly and likely decreased the possibility for coal fines accumulation and jamming. The oxidant volume and its contacting time with coal are likely to be important factors. More oxidant and/or longer contacting time might result in over stimulation, because the higher degree of oxidation could soften the coal and destroy the coal structural integrity. Furthermore, it could generate coal fines which, in addition to minerals, could migrate, accumulate and jam in the fractures. Optimising the oxidant volume and/or the flooding time is expected to be difficult without having detail on the coal properties. In this study, the majority of the new fractures and the void space were produced in the bright coal bands, but also in the dull coal section for coal BH3. The significant change that occurred in the bright coal in this study conflicts with our previous finding that NaClO prefers to attack the dull coal rather than the bright coal when it flowed through the artificial cleats across both bright coal and dull coal (Jing et al., 2018b). To
5. Conclusion To enhance coal seam permeability, acid stimulation (1% HCl) and oxidant stimulation (1% NaClO) were both examined. The core flooding tests could measure the permeability variation during the stimulation procedure, while the CT scanning technique allows the comparison of the coal structural change after stimulation. The effluent from the core flooding test was collected and analysed for elemental composition to examine the chemical reactions happened during the stimulation procedure. The results have led to the following conclusions: 1. Acid stimulation can enhance the horizontal coal permeability through demineralisation evidenced by CT scanning observations and higher element concentrations including Ca, Fe, Al in the effluent. 2. NaClO oxidation can widen coal fracture apertures and generate horizontal fractures and/or increase the void space. These changes are caused primarily by chemical attack, confirmed by the high DOC concentration in the effluent. 3. Despite increasing the void space, oxidant stimulation decreased the vertical permeability. It is proposed that oxidant attack may soften the coal, causing it to collapse in on the newly formed void space due to confining pressure and leading to a drop in permeability. The confining pressure arrangement used here may not represent the confining pressure arrangement downhole. 4. Oxidant stimulation may increase permeability where the mineralogy acts as a support against confining pressure. 5. Based on the samples in this study, it could be proposed that NaClO oxidation is lithotype independent and attacks the coal section that is most porous or contains pre-existing fractures. This study shows the potential of both acid stimulation and oxidant stimulation to increase coal seam permeability. Some uncertainties still exist such as the stimulant contact time and response to confining pressures. The differences in the rank of the samples or the organic petrology may significantly affect the stimulation results on coal permeability and is the subject of ongoing research. Declaration of Competing Interests The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgement We gratefully acknowledge the aid of Dr. Gang Xia, School of Earth and Environmental Science, for the coal sample preparation. We also acknowledge Dr. Karine Mardon at The Centre for Advanced Imaging, for X-ray micro-CT scanning, and the facilities and technical assistance at the Centre for Microscopy & Microanalysis, University of Queensland. This work was supported by the Centre for Coal Seam Gas, The University of Queensland and its industry partners APLNG, Arrow Energy, QGC and Santos as well as the China Scholarship Council. The editor Prof Ralf Littke and the reviewer Dr. Tim A. Moore are highly 10
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
appreciated for their comments.
Laubach, S., Marrett, R., Olson, J., Scott, A., 1998. Characteristics and origins of coal cleat: a review. Int. J. Coal Geol. 35, 175–207. Liu, S., Harpalani, S., 2013. Permeability prediction of coalbed methane reservoirs during primary depletion. Int. J. Coal Geol. 113, 1–10. Liu, F.-J., Wei, X.-Y., Zhu, Y., Gui, J., Wang, Y.-G., Fan, X., Zhao, Y.-P., Zong, Z.-M., Zhao, W., 2013a. Investigation on structural features of Shengli lignite through oxidation under mild conditions. Fuel 109, 316–324. Liu, F.-J., Wei, X.-Y., Zhu, Y., Wang, Y.-G., Li, P., Fan, X., Zhao, Y.-P., Zong, Z.-M., Zhao, W., Wei, Y.-B., 2013b. Oxidation of Shengli lignite with aqueous sodium hypochlorite promoted by pretreatment with aqueous hydrogen peroxide. Fuel 111, 211–215. Mayo, F.R., 1975. Application of sodium hypochlorite oxidations to the structure of coal. Fuel 54, 273–275. Moore, T.A., 2012. Coalbed methane: a review. Int. J. Coal Geol. 101, 36–81. Palmer, I., 2010. Coalbed methane completions: a world view. Int. J. Coal Geol. 82, 184–195. Palmer, I., Mavor, M., Seidle, J., Spitler, J., Voiz, R., 1993. Openhole cavity completions in coalbed methane wells in the San Juan Basin. J. Pet. Technol. 45, 1,072–071,080. Ramandi, H.L., Liu, M., Tadbiri, S., Mostaghimi, P., 2018. Impact of dissolution of syngenetic and epigenetic minerals on coal permeability. Chem. Geol. 486, 31–39. Turner, L.G., Steel, K.M., 2016. A study into the effect of cleat demineralisation by hydrochloric acid on the permeability of coal. J. Nat. Gas Sci. Eng. 36, 931–942. Underschultz, J., 2016. Chapter 28 Unconventional Gas. In: Devasahayam, Sheila, Dowling, Kim, Mahapatra, Manoj (Eds.), Sustainability in the Mineral/Energy Sectors. CRC Press. Vasyuchkov, Y.F., 1985. A study of porosity, permeability, and gas release of coal as it is saturated with water and acid solutions. J. Min. Sci. 21, 81–88. Yao, Z.-S., Wei, X.-Y., Lv, J., Liu, F.-J., Huang, Y.-G., Xu, J.-J., Chen, F.-J., Huang, Y., Li, Y., Lu, Y., 2010. Oxidation of Shenfu coal with RuO4 and NaOCl. Energy Fuel 24, 1801–1808. Zou, Y., Zhang, S., Zhang, J., 2014. Experimental method to simulate coal fines migration and coal fines aggregation prevention in the hydraulic fracture. Transp. Porous Media 101, 17–34.
References Balucan, R.D., Turner, L.G., Steel, K.M., 2016. Acid-induced mineral alteration and its influence on the permeability and compressibility of coal. J. Nat. Gas Sci. Eng. 33, 973–987. Balucan, R.D., Turner, L.G., Steel, K.M., 2018. X-ray μCT investigations of the effects of cleat demineralization by HCl acidizing on coal permeability. J. Nat. Gas Sci. Eng. 55, 206–218. Colmenares, L.B., Zoback, M.D., 2007. Hydraulic fracturing and wellbore completion of coalbed methane wells in the Powder River Basin, Wyoming: implications for water and gas production. AAPG Bull. 91, 51–67. Gamson, P., Beamish, B., Johnson, D., 1996. Coal microstructure and secondary mineralization: their effect on methane recovery. Geol. Soc. Lond., Spec. Publ. 109, 165–179. Jing, Z., 2019. Oxidant Stimulation of Coal Seams to Increase Coal Seam Permeability. The University of Queensland, Brisbane. Jing, Z., Balucan, R.D., Underschultz, J.R., Steel, K.M., 2018a. Oxidant stimulation for enhancing coal seam permeability: swelling and solubilisation behaviour of unconfined coal particles in oxidants. Fuel 221, 320–328. Jing, Z., Mahoney, S.A., Rodrigues, S., Balucan, R.D., Underschultz, J., Esterle, J.S., Rufford, T.E., Steel, K.M., 2018b. A preliminary study of oxidant stimulation for enhancing coal seam permeability: effects of sodium hypochlorite oxidation on subbituminous and bituminous Australian coals. Int. J. Coal Geol. 200, 36–44. Johnson, R., Glassborow, B., Meyer, J., Scott, M., Datey, A., Pallikathekathil, Z., 2010. Utilising current technologies to understand permeability, stress azimuths and magnitudes and their impact on hydraulic fracturing success. APPEA J. 50, 736. Kumar, H., Lester, E., Kingman, S., Bourne, R., Avila, C., Jones, A., Robinson, J., Halleck, P.M., Mathews, J.P., 2011. Inducing fractures and increasing cleat apertures in a bituminous coal under isotropic stress via application of microwave energy. Int. J. Coal Geol. 88, 75–82.
11
Contents lists available at ScienceDirect
International Journal of Coal Geology journal homepage: www.elsevier.com/locate/coal
Chemical stimulation for enhancing coal seam permeability: Laboratory study into permeability variation and coal structure examination Zhenhua Jinga,b, Reydick D. Balucanb, Jim R. Underschultzc, Songqi Pand, Karen M. Steelb,
T
⁎
a
School of Earth and Space Sciences, Peking University, Beijing 100871, China School of Chemical Engineering, The University of Queensland, St Lucia 4072, Australia c Centre for Natural Gas, The University of Queensland, St Lucia 4072, Australia d Research Institute of Petroleum Exploration and Development, PetroChina, Beijing, 100083, China b
A R T I C LE I N FO
A B S T R A C T
Keywords: Coal seam gas Permeability Acid stimulation HCl Oxidant stimulation NaClO
Many potential coalbeds have sub-economic permeability for coal seam gas (CSG) extraction even though all the other key characteristics (gas content, thickness etc.) can meet the requirement for successful production. To enhance coal seam permeability various CSG stimulation techniques, including hydraulic fracturing, cavity well completions and horizontal wells, have been used, but none is ubiquitously successful. The potential of chemical stimulation including both acid (1% HCl) and oxidant (1% NaClO) stimulation are examined in this paper as an alternative method to enhance coal seam permeability. Acid stimulation targets the removal of certain minerals in coal cleats, while the oxidant dissolves the coal matrix, thereby etching pre-existing coal cleats and perhaps forming new ones. The permeability variation during stimulations are measured with core flooding tests and the corresponding coal structural changes are examined using X-ray microcomputed tomography (μCT) technique. The chemical reaction mechanisms are confirmed by elemental analysis via inductively coupled plasma optical emission spectroscopy (ICP-OES) and dissolved organic carbon (DOC) in the effluent. Acid stimulation was performed horizontally with cube samples cut from coal cores (Bowen Basin, Australia) and it exhibited a positive effect on coal permeability, which is proposed to result from demineralisation as shown by before and after CT scanning observations together with high calcium (Ca) content in the effluent. NaClO oxidation was able to etch pre-existing coal cleat surfaces, widen cleat apertures and generate new horizontal fractures and/or void space. These changes were caused primarily by chemical attack, confirmed by the high DOC concentration in the effluent. Oxidant stimulation caused a decrease in vertical permeability and an increase in horizontal permeability for some cores. The decrease in permeability is proposed to be due to a combination of increased void space weakening the coal associated with movement of the coal into newly created space against confining pressure. Minerals in the cleats appear to play a role keeping void space open and preventing collapse under confining pressure. Furthermore, the results for these samples indicate that NaClO oxidation appears to be lithotype independent and preferentially attacks coal sections that have more initial pores or fractures.
1. Introduction Commercial coal seam gas (CSG) extraction has been well established in a number of countries throughout the world, including the USA, Australia, China, India and Canada (Underschultz, 2016). The permeability of coal is a key controlling factor for gas migration in coal reservoirs and is often a critical technical barrier preventing economically viable gas production (Kumar et al., 2011; Moore, 2012). Many low permeability CSG reservoirs require stimulation to allow economic development, even if the seams contain high gas contents (Colmenares
⁎
and Zoback, 2007; Gamson et al., 1996; Laubach et al., 1998; Liu and Harpalani, 2013; Moore, 2012). Although various CSG stimulation techniques, including hydraulic fracturing, cavity well completions and horizontal wells (Colmenares and Zoback, 2007; Palmer, 2010; Palmer et al., 1993), have been used commercially, their success in some wells remain unsuccessful (Johnson et al., 2010; Palmer et al., 1993; Zou et al., 2014). Alternative stimulation options are desirable for increasing gas extraction from coal seams. Chemical stimulation has recently been studied, including both acid and oxidant stimulation (Balucan et al., 2016; Jing et al., 2018a; Turner
Corresponding author. E-mail address: [email protected] (K.M. Steel).
https://doi.org/10.1016/j.coal.2019.103375 Received 16 September 2019; Received in revised form 18 December 2019; Accepted 18 December 2019 Available online 20 December 2019 0166-5162/ © 2019 Published by Elsevier B.V.
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
enabled us to identify the coal structure modification and facilitate understanding of different stimulation effects and mechanisms.
and Steel, 2016). The former targets the removal of minerals from the cleats, while the latter is directed at dissolving the coal matrix to etch the existing cleat surfaces and increase cleat apertures and/or generate new fractures (Jing, 2019). In addition, weakening the coal matrix by oxidant reactions may also prove applicable as a pre-treatment to enhance hydraulic fracturing performance. For low permeability coal seams with highly mineralised fractures, dissolution of mineral occlusions will likely increase fluid flow capacities (Balucan et al., 2016; Turner and Steel, 2016). Vasyuchkov (1985)reported aqueous HCl (2, 4 and 6%) solutions could enhance the coal permeability dramatically from less than 0.1 mD to over 15.1 mD. Turner and Steel (2016) conducted core flooding tests with coal samples containing various cleat and mineralogical properties. The resulting permeability enhancement was “mineral type and rate of occurence” dependent that the largest increases in permeability were achieved where cleats were rich in carbonates, however, the presence of clays placed limitations on the extent of enhancement. Balucan et al. (2018) investigated the acid-induced physical and chemical changes in core samples to gain improved understanding of permeability enhancement in the vertical flow direction. There was clear evidence for cleat opening via mineral dissolution as well as mineral mobilizationaccumulation. Based on the X-ray microcomputed tomography (μCT) image analyses, they conducted flow simulation experiments and reported that acidizing could enhance permeability not only in the vertical direction but also in the lateral or horizontal direction (Balucan et al., 2018). Similarly, Ramandi et al. (2018) applied the μCT and numerical simulation techniques to investigate the effect of dissolution of syngenetic and epigenetic minerals on coal permeability and reported that coal permeability increase depends significantly on epigenetic mineral dissolution, and secondarily on syngenetic mineral dissolution. Although the above studies have examined the effect of acidization on coal permeability experimentally and numerically, uncertainty still remains. For example, few of their results relate to the horizontal permeability. Additionally, in the core flooding tests conducted by Turner and Steel (2016), the HCl was injected in the same direction with brine. However, to mimic how stimulation would be conducted downhole, forward flow to characterise initial permeability followed by reverse flow with acid to simulate the downhole treatment and finally forward flow again to characterise the final permeability is required. In addition to acid stimulation, we have recently reported on oxidant stimulation of coal seams (Jing et al., 2018a,b) where NaClO was determined to be one of the most effective oxidants, given its high reactivity and soluble reaction products (Liu et al., 2013a,b; Yao et al., 2010). The action of NaClO in artificial cleats on natural coal surfaces revealed increases in the cleat aperture (Jing et al., 2018b). Furthermore, NaClO oxidation appeared to increase the coal porosity by enlarging the pore sizes, which was measured by mercury intrusion porosimetry and observed by Scanning Electron Microscopy (SEM) imaging (Jing et al., 2018b). The matrix pore size enlargement and the cleat aperture dilation are proposed to be capable to increase the cleat system connectivity, thus increasing coal seam permeability. In this paper, coal core samples containing fully developed cleat systems were used to explore the effect of acidisation and oxidation on coal permeability under confining pressure. The permeability was measured using the core flooding rig which enabled the forward flow followed by a reverse flow and a final forward flow again to mimic how stimulation would be conducted downhole. The cubic samples cut from the cores allowed us to measure horizontal permeability variation. In terms of the oxidant stimulation, previously we have reported on a screening procedure which identified NaClO (Jing et al., 2018a) as the most promising oxidant and examined its effect on coal structures including pores and cleats (Jing et al., 2018b), Here we report on the coal permeability variation after NaClO treatment, both vertically and horizontally, to investigate the efficacy of NaClO to increase coal permeability. Furthermore, CT scanning of samples before and after treatment
2. Methodology 2.1. Coal samples The chemical stimulation in this study included acid stimulation with 1% HCl and oxidant stimulation with 1% NaClO. Stimulation was examined perpendicular and parallel to the coal bedding plane. The vertical permeability was studied on coal cores (76 mm diameter) collected from Bowen Basin (labelled BV), while the horizontal permeability was measured with cube samples (40 × 40 × 40 mm) which were cut from cores from the same basin but different boreholes (labelled BH). Core sample information including sample size and mineral information has been previously reported (Turner and Steel, 2016). The two coals shared similar vitrinite reflectance that was 0.86% for BV coal and 0.84% for BH coal. Their mineral compositions were slightly different in that BV coal was primarily comprised of kaolinite while BH coal was mainly composed of carbonates (calcite, siderite) and barite and secondarily of kaolinite. In addition, BV coal and BH coal both contain dull and bright coal bands as evidenced by visual observation and CT scanning. The acid stimulation for coal cores BV1 and BV2 had been conducted with 1% HCl previously where they were labelled “A1” and “A2”, respectively (Turner and Steel, 2016). In this work, these cores have been treated with NaClO. Cube samples (BH samples) were taken from the same borehole as the samples labelled “C” in previous research (Turner and Steel, 2016). In this work, these cores were treated with HCl and NaClO. 2.2. Core flooding test Steady state core flood permeability testing was performed using the core flooding rig system(Fig. 1) and it was updated from the rig previously reported by Turner and Steel (2016). The system mainly includes a Hassler core holder; one high-pressure Quizix QX6000 dual cylinder pump for fluid injection; one Redox gas/oil tank for confining pressure control; Gems 3200 series pressure transducers (0–100 bar) to measure both pore pressure drop and confining pressure; and a domeloaded back pressure regulator to maintain fluid pressure drop across the sample. Wetted parts were predominantly Hastelloy to provide superior corrosion resistance to brine, dilute acids and oxidants. The components in the system have been reported previously (Balucan et al., 2016; Turner and Steel, 2016). Instead of flooding brine and the stimulants in the same direction, the case in previous research by Turner and Steel (2016), this system allowed the stimulant to flow in a reverse direction to mimic the stimulation procedure in-situ. Brine and acid were injected through the sample by the pump, while compressed air was used to push oxidant into the coal samples in order to protect the pump from oxidation. For each run, the oxidant volume flowing through the coal was deliberately monitored to be less than the oxidant cylinder volume to prevent gas from entering the coal samples. Prior to the flooding test, the coal samples were X-ray CT scanned to characterise the coal structures, particularly the fracture systems. The flooding tests start with the coal samples saturated in brine (4% KCl) to enable the coal to be at equilibrium with in-situ conditions and consolidated via repeated loading/unloading cycles. Once the saturation was established, a forward brine injection was conducted until the permeability measurement stabilised, which established the initial permeability baseline. Afterwards, the stimulant (either 1% HCl or 1% NaClO) was injected in the opposite “reverse” direction until the permeability stabilised under the same circumstances, i.e. same confining pressure (Pc), inlet pressure (P1), outlet pressure (P2) and effective pressure (Pe). This was followed by a second forward flow with brine to compare the permeability change before and after stimulation. 2
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
Fig. 1. Schematic of core flooding system. Table 1 Pressure system of the coal flooding test for each sample. Sample
BV1 BV2 BH1 BH2 BH3
Stimulant
1% 1% 1% 1% 1% 1% 1%
NaClO & 4% KCl NaClO & 4% KCl HCl & 4% KCl NaClO & 4% KCl HCl & 4% KCl NaClO & 4% KCl NaClO & 4% KCl
Flooding direction
Pressure system
Vertical Vertical Horizontal Horizontal Horizontal Horizontal Horizontal
P1 (bar)
P2 (bar)
Pc (bar)
Pe (bar)
10 10 15 10 15 10 10
4 4 4 4 4 4 4
25 25 25 25 25 25 50
18 18 15.5 18 15.5 18 43
Image segmentation into three coal components, namely unfilled fracture (voids or fracture porosity), coal matrix and minerals, was achieved via greyscale unit (GU) thresholding. Components were allocated specific greyscale ranges following detailed surveys of the greyscale values of visually discernible structures. Fracture or mineral fractions (Xf) of individual slices were estimated using Eq. (1), where ∑PXGUrange is the sum of pixels within the greyscale range (lower to upper limit) of the component and ∑PX0−255 is the total pixels for the entire slice. Following estimates validation, the core profiles were compared to identify the structural changes of coal structures.
Subsequently, the sample was X-ray CT scanned again to investigate the coal structural change after the flooding test. The pressure details about the core flooding tests are summarised in Table 1. Liquid samples were regularly extracted from the core holder outlet and analysed for element concentrations via inductively coupled plasma optical emission spectroscopy (ICP-OES) and for dissolved organic carbon (DOC). The procedure for ICP-OES and DOC analysis has been previously described (Jing et al., 2018a; Turner and Steel, 2016). 2.3. X-ray computed tomography (CT) scanning An Inveon Multimodality PET/CT scanner (Seimens) at the Centre for Advanced Imaging (CAI), The University of Queensland, was used (voltage = 80 kV, beam current = 0.5 mA) to image the coal sample before and after stimulation. Scanned images of the core were reconstructed, visualised and normalised using the Inveon™ Research Workplace software (Seimens IRW v4.2). The 3D reconstructions and visualisation enabled qualitative surveys. These surveys were fundamental to understanding core structure that is essential for validating the generated core profiles. Greyscale image normalisation, attained by keeping the contrast parameters constant, allowed for comparison of the images before and after stimulation. These axial greyscale images (display resolution = 44 μm) were then used for mapping the mineral and fracture profiles of the core before and after stimulation. Core profiling involves image stack generation, slice histogram extraction, image segmentation and fractionation. The image stacks each containing ~1600 greyscale axial slices (images) were manually registered, stabilised and cropped prior to extracting the slice histograms using plugins and macros embedded in ImageJ software.
Xf =
∑ PXGUrange ∑ PX0 − 255
(1)
3. Results 3.1. Effect of oxidant stimulation on vertical permeability Two coal cores, BV1 and BV2, which have been previously treated with 1% HCl were used to examine the effect of oxidation on vertical permeability. 3.1.1. Coal core sample BV1 Coal BV1 was initially flooded perpendicular to the bedding plane by 4% KCl (forward flow) until the permeability reached a plateau and then 1% NaClO diluted with 4% KCl was injected in the opposite direction (reverse flow) to stimulate the coal until the permeability stabilised. Afterwards, a second forward flow with 4% KCl was conducted to examine the permeability change by comparing with the initial 3
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
same site of the upper coal core before and after oxidation, while Fig. 5c and g are from the same position of the lower section. The apparent change could be observed by comparing Fig. 5b and f. After oxidation, the void space in Fig. 5f accounted for 30% of the whole image and a horizontal fracture existed on the left of the image. In the lower section, by contrast, the structure primarily kept intact without discernible change (Fig. 5c and g). The different effects of oxidation on upper and lower sections can be further demonstrated by the VF distribution (Fig. 5h), where the VF in the upper section after oxidation (green line) is considerately higher than the initial VF (Blue line), while in the lower section the two lines overlapped. 3.2. Effect of acid and oxidant stimulation on horizontal permeability
Fig. 2. Permeability variation of BV1 in 1% NaClO stimulation. P1 = 10 bar, P2 = 4 bar, Pc = 25 bar.
Cube samples enable the measurement of horizontal permeability parallel to the coal bedding plane, which reflects the in-situ permeability direction parallel to bedding and perpendicular to the borehole which is the desirable direction to examine the permeability for a wellbore stimulation effect. Few researchers have reported the effect of acid stimulation on horizontal permeability and even less the effect of oxidation. In this study three cube samples were used and two of them (BH1 and BH2) experienced acid stimulation first and oxidant stimulation afterwards, while the other one (BH3) underwent oxidant stimulation only.
forward flow. The permeability variation during this procedure was illustrated in Fig. 2 for coal BV1. The initial stable permeability with 4% KCl was 0.59 mD and gradually dropped to 0.02 mD during the injection of 230 mL NaClO. Afterwards, the forward permeability post oxidation stabilised at 0.14 mD. To understand the permeability decline, the coal core structural change induced by oxidation was examined based on CT scanning data as shown in Fig. 3 where the core flooding was directed from top to bottom. Compared with the initial coal core 3D image (Fig. 3a), NaClO oxidation induced a group of fractures occurring at the middle of the coal core (Fig. 3e). The cross-section CT images could further illustrate the fracture pattern from a top to bottom view. On Fig. 3f, the fractures or void spaces, which did not exist pre oxidation (Fig. 3b), extended horizontally through the coal core. The void space fraction (VF) after oxidation increased significantly, particularly in the middle location of the coal (Fig. 3g) where it was initially more porous than other parts in the core (Fig. 3c). In the same section, the 3D image of the oxidised coal core revealed a roughly horizontal connected layer as shown by the blue layer in Fig. 3h.
3.2.1. Cube sample BH1 Acid stimulation with 1% HCl and 4% KCl was firstly conducted for coal BH1. The corresponding permeability variation and the element concentrations in the effluent are shown in Fig. 6. The permeability of initial forward flow was 0.033 mD. When 1% HCl was flooded in the reverse direction, its permeability increased to 1.2 mD but then gradually declined to 0.38 mD with a total 1550 mL 1% HCl flowing through. The following forward flow started with a permeability of 0.8 mD and eventually reached 0.35 mD, which was more than 10 times higher than the initial forward permeability (0.033 mD). The increase of permeability during HCl flooding was accompanied by a drastic elevation in the calcium (Ca) concentration and a minor iron (Fe) increase as shown in Fig. 6, indicating the significant dissolution of calcite (CaCO3) and a slight dissolution of ankerite (Ca(Fe, Mg, Mn)(CO3)2). The concentration of aluminium (Al) and barium (Ba) were quite low. After 1% HCl flooding, the coal was subjected to oxidant stimulation with 1% NaClO and Fig. 7 shows the permeability and DOC concentration variation during this procedure. The initial forward permeability was 0.038 mD. Note that the initial forward permeability in 4% KCl (F1) was not equal to the final forward flow (F2) in the acid stimulation because they did not share the same pressure conditions (Table 1). The reverse permeability during oxidant injection started at 0.118 mD and generally dropped to 0.04 mD after 60 mL 1% NaClO flowing through. The forward flow post oxidation showed a horizontal permeability of 0.003 mD, revealing the damage to coal flow capacity caused by oxidation. In Fig. 7, the DOC concentration in the initial forward flow with 4% KCl was 14 mg/L, and drastically increased with the injection of NaClO to 350 mg/L. Afterwards, it dropped to 26.62 mg/L in the second brine forward flow. The high DOC concentration in the effluent during NaClO injection demonstrated the chemical attack happening to the coal matrix. Fig. 8 shows the coal structural change associated with both acid and oxidant stimulation based on CT scanning images. The CT image slides were collected and shown in two directions: Y direction, which is the horizontal flooding direction; Z direction, which is the top to bottom vertical direction. In the Y direction, the CT images from the same site of the coal pre stimulation, post acid stimulation and post oxidant stimulation are shown in Fig. 8b, c and d, respectively. The sites
3.1.2. Coal core sample BV2 The second core sample (BV2) was pre-treated with 30% H2O2 because H2O2 has been reported to be able to enhance the oxidative effect of NaClO for coals (Mayo, 1975). Coal BV2 experienced two stages of oxidation, 30% H2O2 and then 1% NaClO. The corresponding permeability variation with each stage is shown in Fig. 4. The permeability of initial forward flow (F1) with brine (4% KCl) was 2.01 mD, followed by the reverse (R1) with 30% H2O2 (total 40 mL), where the permeability dropped to 0.08 mD. A second forward flow (F2) was conducted with a measured permeability of 0.59 mD, showing the detrimental effect of 30% H2O2 on permeability. Afterwards, 50 mL of 1% NaClO was injected into the same sample, and the permeability was recorded to steadily decline from 0.25 mD to 0.05 mD. Subsequently a final forward flow (F3) showed the permeability stabilised at 0.17 mD, much lower than the initial 2.01 mD. Dissolved organic carbon (DOC) dissolved in the effluent solutions showed the concentration in brine was low at 23.5 mg/L (Fig. 4). During the H2O2 flooding, DOC increased slightly to 31.8 mg/L. DOC concentration in the effluent increased with injection of NaClO. It then dropped during the final brine flooding. The structural variation of sample BV2 before and after treatment with 30% H2O2 and 1% NaClO are shown in Fig. 5. The coal core was divided into upper and lower sections (Fig. 5a). In the upper section, more void spaces existed initially, as shown in Fig. 5d illustrating the void space fraction (VF) distribution along the coal core axis, while the lower section was physically tighter. After oxidation treatment, massive coal structural change occurred in the upper section, rather than its lower counterpart (Fig. 5e). This could be clearly illustrated by the cross-section images. Fig. 5b and f are the cross-section CT images at the 4
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
Fig. 3. BV1 core structure analysis based on X-ray μCT scanning. a. Vertical cross-section across core length before oxidation; b. horizontal slice through image shown in (a); c. void space fraction along core length before oxidation; d. 3D image of mineral and void space in before oxidation; e. Vertical cross-section across core length after oxidation; f. horizontal slice through image after oxidation at the same location as that shown in (b); g. void space fraction along core length after oxidation; h. 3D image of mineral and void space in the core after oxidation;
(Fig. 8b) vanished after acid stimulation (Fig. 8c). Similarly, the contents of minerals in the dull bands could be observed to decrease albeit less obviously. Then after oxidant flooding, the site shows as a vacant fracture (Fig. 8d). The variation of fracture features explicitly indicates the demineralisation effect during acid stimulation and the etching effect of oxidant on fractures. These fracture variations were confirmed by viewing CT images in the Z direction as shown in Fig. 8e (pre-acid), Fig. 8f (post-acid) and Fig. 8g (post-oxidant). The images are a cross section selected in the bright coal band (red rectangle in Fig. 8a), which was initially highly mineralised (Fig. 8e). The minerals filling in the cleats highlighted by the red ellipse were removed after acid stimulation and a void cleat appeared at the same section after oxidation. Moreover, after oxidation, the vacant cleats formed a connected fracture system across the sample. Fig. 8h, i, g represent the 3D image of both void space and minerals in the coal in the Z direction (viewing from top to bottom in Fig. 8a), before stimulation, post acid and post oxidant stimulation, respectively. The minerals were extensively distributed in the whole sample before acid stimulation, but after acid flooding, the amount of minerals decreased, especially in the middle of the sample. After oxidation, an
Fig. 4. Permeability and DOC concentration variation at different oxidation stages for coal BV2. P1 = 10 bar, P2 = 4 bar, Pc = 25 bar.
selected by the red dash rectangle in these figures exemplify a typical change of the coal structure undergoing the stimulations. The minerals initially filled in the fractures in the bright coal band before stimulation 5
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
Fig. 5. BV2 sample coal structure analysis based on core CT scanning. a. 3D image before oxidation; b. horizontal cross-sectional slice in upper section before oxidation; c. horizontal cross-sectional slice in lower section before oxidation; d. void space fraction before oxidation; e. 3D image after oxidation; f. cross-sectional slice in same spot as (b) after oxidation; g. cross-sectional slice in same spot as (c) after oxidation; h. comparison of void space and mineral fraction before and after oxidation. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)
may be the vacant cleat not filled by minerals. The initial mineral distribution was reflected in Fig. 9b (blue line) with two humps showing the mineral fractions, while the initial void space was shown in Fig. 9c (blue line). After acid stimulation, the drop of mineral fraction is clearly shown in Fig. 9b where the original two humps disappeared and the mineral fraction stayed at a similar low level along the whole coal sample. This means most of the minerals were removed by 1% HCl. By comparison, the void space created is also shown as two humps at the locations where minerals existed initially (red line Fig. 9c). Afterwards, the oxidation further enlarged the void space regardless of the coal type, as shown in Fig. 9c. NaClO appeared to barely influence the minerals. Overall, the results clearly quantified the demineralisation of acid stimulation and the enlargement of void space after oxidation. However, the increase of void space after oxidation detected by CT imaging could not explain the bulk decline of permeability.
Fig. 6. Permeability and element concentration change during 1% HCl stimulation for coal BH1. Ca: Calcium; Fe: Iron; Ba: Barium; Al: Aluminium; P: Phosphorus. P1 = 15 bar, P2 = 4 bar, Pc = 25 bar.
3.2.2. Cube sample BH2 Sample BH2 shares a similar lithotype with BH1 that contains both dull coal and bright coal layers. The same stimulation procedures were conducted to BH2, i.e. acid stimulation first followed by oxidant stimulation. Fig. 10 illustrates the permeability variation and the element concentration in the flooding effluent for coal BH2 during 1% HCl stimulation. The initial forward flow had a stable permeability of 0.35 mD, which generally increased to 1.21 mD after 1700 mL 1% HCl flowed through. Afterwards, the post-stimulation forward flow stabilised at 0.85 mD. The permeability variation shows the 1% HCl stimulation could elevate the permeability 2.42 times for coal BH2. Upon injection of HCl, there was a spike in the concentration of Ca in the effluent, from 65 ppm in the initial brine flow to 500 ppm in the 1% HCl flooding. Then it decreased dramatically with increasing HCl injection presumably due to complete removal of accessible calcite (CaCO3). The concentrations of P and Fe were significant, indicating dissolution of apatite (Ca5(PO4)3(F,Cl,OH)) and ankerite (Ca(Fe, Mg, Mn)(CO3)2). Concentration of Ba mainly derived from barite (BaSO4) was much higher in the initial brine flooding for coal BH2 than that for coal BH1, indicating a high barite content existing in the coal structures. Barite was observed by SEM in the coal cleats after acid stimulation, as shown in Fig. 11. Dissolution of apatite could produce weak HF which could then
Fig. 7. Permeability variation during 1% NaClO stimulation for coal BH1. P1 = 10 bar, P2 = 4 bar, Pc = 25 bar.
apparent void fracture could be observed across the whole sample in the XY plane (horizontal plane). Additionally, the void space fraction (VF) and mineral fraction (MF) change along the coal sample from top to bottom (z direction) further demonstrating the demineralisation effect of acid stimulation and the etching effect of oxidant stimulation as shown in Fig. 9. Fig. 9a illustrates two distinct regions of which the upper part is dull coal and the lower part appeared to be bright coal filled with cleats. Some void space (shown as blue colour in Fig. 9a) could be noted in the bright coal and it 6
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
Fig. 8. BH1 coal structure change after acid and oxidant stimulation. a. A diagram of coal BH1 showing the stimulation direction is along Y; b, c and d show the same image viewed in the stimulation direction before and after acid and oxidant stimulation. e, f and g show images of the cross-section shown in a; h, i and j show 3D images for coal BH1 before and after stimulations from a view angle in z direction.
After acid stimulation, a forward flow was conducted with an inlet pressure reduced to 10 bar to build the permeability baseline for oxidant stimulation, which was measured as 0.14 mD. Then the 1% NaClO flooding followed, where the reverse direction permeability reached quite high at the beginning (7.5 mD) and declined to 2.3 mD with a
dissolve kaolinite (Al2Si2O5(OH)4)). Significant levels of Al were observed at the end of acid flooding. CT scanning images failed to explain the coal BH2 structural change due to the X-ray beam hardening caused by extensive mineral content, particular the high concentration of barite. 7
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
Fig. 9. Void space fraction and mineral fraction for coal BH1. a. 3D image for coal BH1 showing mineral and void space before oxidation and the flooding direction was along Y; b. Mineral fraction before and after acid and oxidant stimulation; c. Void space fraction before and after acid and oxidant stimulation. (For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)
Fig. 10. Permeability variation in 1% HCl stimulation for coal H2. Ca: Calcium; Fe: Iron; Ba: Barium; Al: Aluminium; P: Phosphorus.P1 = 15 bar, P2 = 4 bar, Pc = 25 bar.
Fig. 12. Permeability variation in 1% NaClO stimulation for coal BH2. P1 = 10 bar, P2 = 4 bar, Pc = 25 bar.
Fig. 11. SEM image of Barite in the cleat of coal BH2 after acid stimulation. a. the Barite on the inlet surface of BH2; b. Barite in the cleat of the inlet of BH2. 8
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
coal to keep its minerals intact. Fig. 13 shows the horizontal permeability variation during the single oxidant stimulation. The initial permeability was 1.95 mD, which dropped to 0.62 mD at the beginning of oxidation, and after 1 hour flowing, the permeability shot up to 8.2 mD and fluctuated around this value. Then the forward permeability of brine was 6.84 mD, 4.38 times higher than the initial (1.95 mD). CT images pre/post oxidation for coal BH3 revealed the coal structural changes caused by oxidation as shown in Fig. 14. Combining the hand sample observation and the 3D image (Fig. 14a), it could be confirmed that there existed multiple layers of bright coal and one dull coal layer with an initial fracture across shown as a blue layer. After oxidation (Fig. 14c), void space proportion could be viewed increasing in the middle of the sample. Fig. 14b and d show the cross-section CT images of the initial fracture in dull coal before and after oxidation, respectively. The comparison of these images could identify two fracture feature changes that could benefit the permeability. The first was the increase of fracture aperture. Before oxidation, the fracture apertures were measured as 134 μm and 126 μm for spot 1 and spot 2, respectively, and they were enlarged to 174 μm and 228 μm after oxidation. The second was the extension of the existing fracture, which was illustrated in Fig. 14d selected by the red rectangle, where a new fracture was generated and connected to the initial fracture.
Fig. 13. Permeability variation during 1% NaClO oxidation for coal BH3. P1 = 10 bar, P2 = 4 bar, Pc = 50 bar.
total flooding volume of 1300 mL. The post-stimulation forward flow started at 3.5 mD and ended with 0.71 mD, which was 5 times higher than the pre-oxidation permeability as shown in Fig. 12. DOC concentration in the initial forward flow was 31 mg/L and shot up to 108 mg/L at the beginning of NaClO injection followed by a gradual decrease. The final brine flow was 19 mg/L. This DOC trend for coal BH2 was different to that for coal BV1 and BH1, where the DOC concentration gradually increased during the NaClO injection. The difference is proposed to be caused by the permeability level during the NaClO flooding, which was significantly higher here than that for coal BV1 and BH1. The higher permeability allows the dissolved organic carbon to flow out immediately, while the lower permeability hampers the flow and delays the detection of dissolved coals. Furthermore, the average DOC concentration for coal BH2 was lower than those for coal BV1 and BH1. This is caused by the significant volume of NaClO flowing through coal BH2, which reduced the average DOC concentration in unit volume (10 mL here).
4. Discussion The horizontal permeability of the two studied cube samples (BH1 and BH2) both increased after acid stimulation. The CT image comparison for coal BH1 before and after oxidation (Fig. 8) showed that minerals were removed in some parts of the cleats after 1% HCl flooding and the mineral fraction (MF) calculation based on CT images confirmed the demineralizing effect. New void space occurred at the locations where initially minerals existed. These generated void spaces may stay open during the flooding test due to the hydraulic pressure and provide pathways for fluid thus elevating the permeability during acid stimulation. The ICP-OES results for the two studied samples confirmed the demineralisation caused by 1% HCl flooding. For both coals, Ca was the main detected element in the effluent, indicating the dissolution of Ca containing minerals including calcite, aparite and ankerite. These results of an increase in horizontal permeability agree with those found
3.2.3. Cube sample BH3 Both of the two cubic samples described above underwent acid stimulation followed by oxidant stimulation. Oxidation showed a detrimental effect on BH1 permeability, but a positive effect on BH2 which contained a significantly higher mineral content, especially with respect to Barite. This may indicate that minerals play a role in oxidant stimulation. Therefore, for coal BH3, only oxidant was flowed through the
Fig. 14. Fracture feature change in coal BH3 after 1% NaClO oxidation. a.3D image for coal BH3 before oxidation where blue shows a large fracture across a dull coal band; b. Cross-section of the large fracture and where aperture was measured at two spots; c. 3D image for coal BH3 after oxidation; d. Cross-section of the large fracture after oxidation. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) 9
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
examine this conflict, the void space fractions for each coal before and after oxidation were analysed and we found that the NaClO attacks in the locations where the void space is highest. The extent of attack appears to be dependent on the surface area. The NaClO oxidation therefore appears to be lithotype independent and the NaClO attacks any coal type as long as it can come to contact with surface area. For example, in coal BH3, where a fracture existed in the dull coal band, coal oxidation not only widened the fracture aperture, but also produced a new fracture.
for vertical permeability for Ca-rich cores (Turner and Steel, 2016), where increases in permeability were observed. After oxidation, the vertical permeability decreased for both coal cores, while the horizontal permeability decreased for BH1, but increased for BH2 and BH3. Analysis of CT scanning images showed three main changes on coal structures after NaClO oxidation. The first is the enlargement of the void space at locations that were initially more porous. The second is the generation of new fractures close to the horizontal bedding direction or the extension of existing fractures. The third is the general enlargement of fracture apertures. These changes could suggest an increase in permeability. For the coal cores BV1 and BV2, a substantial decrease in vertical permeability was observed. It is proposed that the bulk vertical permeability is determined by the section having the lowest permeability. Furthermore, it is proposed that oxidant stimulation weakened/softened the cores. Against a confining pressure, it is proposed that the softened coal collapsed in on the newly formed void spaces, leading to the drop in permeability. For coal cores BH1 and BH2 a decrease and increase in permeability was observed. The mineral difference between BH1 and BH2 may have led to the differences in permeability after stimulation. BH1 and BH2 both experienced 1% HCl flooding, which dissolved a large proportion of minerals. During oxidant flooding, although coal fractures in BH1 were etched and enlarged, the confining pressure may have caused the closure of existing fractures with the decline in fluid pressure (Balucan et al., 2016; Kumar et al., 2011) and even result in the fractures collapsing. It is worthwhile to note that confining pressure is applied from all sides in the set-up (Fig. 1), which may not adequately represent the stress conditions downhole. Therefore, it is uncertain whether collapse in enlarged fractures would occur. For BH2 a significant amount of carbonate minerals were also dissolved and a considerable amount of barite remained. This barite may have helped to maintain the coal structure integrity (i.e. prop open fractures against the confining pressure conditions). The large amount of barite could also support the cleats during the oxidant flooding (Fig. 11). To examine this hypothesis, coal BH3 was directly flooded by oxidant without a pre-treatment with acid to keep the minerals intact. The permeability for BH3 increased from 1.95 mD to 6.84 mD (Fig. 13). The void space proportion in the bright layers was confirmed to be increased (Fig. 14) and the remaining minerals may have protected the cleats from collapsing. Besides the minerals, other factors could affect the permeability during oxidation including specific coal fracture structures, the oxidant volume, the contacting time and pressure. For coal BH1, although the flooding direction was parallel to the face cleats, the face cleats terminated in the middle of the coal and connected with another face cleat by a butt cleat. The connection points with a narrow aperture might act as a restriction point. Coal fines could accumulate here and halt the permeability enhancement. For coal BH3, the face cleats went through the whole sample, making it easier for fluid to flow through without fines jamming. In addition, coal BH3 contained a large fracture which could allow the fluid to flow quickly and likely decreased the possibility for coal fines accumulation and jamming. The oxidant volume and its contacting time with coal are likely to be important factors. More oxidant and/or longer contacting time might result in over stimulation, because the higher degree of oxidation could soften the coal and destroy the coal structural integrity. Furthermore, it could generate coal fines which, in addition to minerals, could migrate, accumulate and jam in the fractures. Optimising the oxidant volume and/or the flooding time is expected to be difficult without having detail on the coal properties. In this study, the majority of the new fractures and the void space were produced in the bright coal bands, but also in the dull coal section for coal BH3. The significant change that occurred in the bright coal in this study conflicts with our previous finding that NaClO prefers to attack the dull coal rather than the bright coal when it flowed through the artificial cleats across both bright coal and dull coal (Jing et al., 2018b). To
5. Conclusion To enhance coal seam permeability, acid stimulation (1% HCl) and oxidant stimulation (1% NaClO) were both examined. The core flooding tests could measure the permeability variation during the stimulation procedure, while the CT scanning technique allows the comparison of the coal structural change after stimulation. The effluent from the core flooding test was collected and analysed for elemental composition to examine the chemical reactions happened during the stimulation procedure. The results have led to the following conclusions: 1. Acid stimulation can enhance the horizontal coal permeability through demineralisation evidenced by CT scanning observations and higher element concentrations including Ca, Fe, Al in the effluent. 2. NaClO oxidation can widen coal fracture apertures and generate horizontal fractures and/or increase the void space. These changes are caused primarily by chemical attack, confirmed by the high DOC concentration in the effluent. 3. Despite increasing the void space, oxidant stimulation decreased the vertical permeability. It is proposed that oxidant attack may soften the coal, causing it to collapse in on the newly formed void space due to confining pressure and leading to a drop in permeability. The confining pressure arrangement used here may not represent the confining pressure arrangement downhole. 4. Oxidant stimulation may increase permeability where the mineralogy acts as a support against confining pressure. 5. Based on the samples in this study, it could be proposed that NaClO oxidation is lithotype independent and attacks the coal section that is most porous or contains pre-existing fractures. This study shows the potential of both acid stimulation and oxidant stimulation to increase coal seam permeability. Some uncertainties still exist such as the stimulant contact time and response to confining pressures. The differences in the rank of the samples or the organic petrology may significantly affect the stimulation results on coal permeability and is the subject of ongoing research. Declaration of Competing Interests The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgement We gratefully acknowledge the aid of Dr. Gang Xia, School of Earth and Environmental Science, for the coal sample preparation. We also acknowledge Dr. Karine Mardon at The Centre for Advanced Imaging, for X-ray micro-CT scanning, and the facilities and technical assistance at the Centre for Microscopy & Microanalysis, University of Queensland. This work was supported by the Centre for Coal Seam Gas, The University of Queensland and its industry partners APLNG, Arrow Energy, QGC and Santos as well as the China Scholarship Council. The editor Prof Ralf Littke and the reviewer Dr. Tim A. Moore are highly 10
International Journal of Coal Geology 219 (2020) 103375
Z. Jing, et al.
appreciated for their comments.
Laubach, S., Marrett, R., Olson, J., Scott, A., 1998. Characteristics and origins of coal cleat: a review. Int. J. Coal Geol. 35, 175–207. Liu, S., Harpalani, S., 2013. Permeability prediction of coalbed methane reservoirs during primary depletion. Int. J. Coal Geol. 113, 1–10. Liu, F.-J., Wei, X.-Y., Zhu, Y., Gui, J., Wang, Y.-G., Fan, X., Zhao, Y.-P., Zong, Z.-M., Zhao, W., 2013a. Investigation on structural features of Shengli lignite through oxidation under mild conditions. Fuel 109, 316–324. Liu, F.-J., Wei, X.-Y., Zhu, Y., Wang, Y.-G., Li, P., Fan, X., Zhao, Y.-P., Zong, Z.-M., Zhao, W., Wei, Y.-B., 2013b. Oxidation of Shengli lignite with aqueous sodium hypochlorite promoted by pretreatment with aqueous hydrogen peroxide. Fuel 111, 211–215. Mayo, F.R., 1975. Application of sodium hypochlorite oxidations to the structure of coal. Fuel 54, 273–275. Moore, T.A., 2012. Coalbed methane: a review. Int. J. Coal Geol. 101, 36–81. Palmer, I., 2010. Coalbed methane completions: a world view. Int. J. Coal Geol. 82, 184–195. Palmer, I., Mavor, M., Seidle, J., Spitler, J., Voiz, R., 1993. Openhole cavity completions in coalbed methane wells in the San Juan Basin. J. Pet. Technol. 45, 1,072–071,080. Ramandi, H.L., Liu, M., Tadbiri, S., Mostaghimi, P., 2018. Impact of dissolution of syngenetic and epigenetic minerals on coal permeability. Chem. Geol. 486, 31–39. Turner, L.G., Steel, K.M., 2016. A study into the effect of cleat demineralisation by hydrochloric acid on the permeability of coal. J. Nat. Gas Sci. Eng. 36, 931–942. Underschultz, J., 2016. Chapter 28 Unconventional Gas. In: Devasahayam, Sheila, Dowling, Kim, Mahapatra, Manoj (Eds.), Sustainability in the Mineral/Energy Sectors. CRC Press. Vasyuchkov, Y.F., 1985. A study of porosity, permeability, and gas release of coal as it is saturated with water and acid solutions. J. Min. Sci. 21, 81–88. Yao, Z.-S., Wei, X.-Y., Lv, J., Liu, F.-J., Huang, Y.-G., Xu, J.-J., Chen, F.-J., Huang, Y., Li, Y., Lu, Y., 2010. Oxidation of Shenfu coal with RuO4 and NaOCl. Energy Fuel 24, 1801–1808. Zou, Y., Zhang, S., Zhang, J., 2014. Experimental method to simulate coal fines migration and coal fines aggregation prevention in the hydraulic fracture. Transp. Porous Media 101, 17–34.
References Balucan, R.D., Turner, L.G., Steel, K.M., 2016. Acid-induced mineral alteration and its influence on the permeability and compressibility of coal. J. Nat. Gas Sci. Eng. 33, 973–987. Balucan, R.D., Turner, L.G., Steel, K.M., 2018. X-ray μCT investigations of the effects of cleat demineralization by HCl acidizing on coal permeability. J. Nat. Gas Sci. Eng. 55, 206–218. Colmenares, L.B., Zoback, M.D., 2007. Hydraulic fracturing and wellbore completion of coalbed methane wells in the Powder River Basin, Wyoming: implications for water and gas production. AAPG Bull. 91, 51–67. Gamson, P., Beamish, B., Johnson, D., 1996. Coal microstructure and secondary mineralization: their effect on methane recovery. Geol. Soc. Lond., Spec. Publ. 109, 165–179. Jing, Z., 2019. Oxidant Stimulation of Coal Seams to Increase Coal Seam Permeability. The University of Queensland, Brisbane. Jing, Z., Balucan, R.D., Underschultz, J.R., Steel, K.M., 2018a. Oxidant stimulation for enhancing coal seam permeability: swelling and solubilisation behaviour of unconfined coal particles in oxidants. Fuel 221, 320–328. Jing, Z., Mahoney, S.A., Rodrigues, S., Balucan, R.D., Underschultz, J., Esterle, J.S., Rufford, T.E., Steel, K.M., 2018b. A preliminary study of oxidant stimulation for enhancing coal seam permeability: effects of sodium hypochlorite oxidation on subbituminous and bituminous Australian coals. Int. J. Coal Geol. 200, 36–44. Johnson, R., Glassborow, B., Meyer, J., Scott, M., Datey, A., Pallikathekathil, Z., 2010. Utilising current technologies to understand permeability, stress azimuths and magnitudes and their impact on hydraulic fracturing success. APPEA J. 50, 736. Kumar, H., Lester, E., Kingman, S., Bourne, R., Avila, C., Jones, A., Robinson, J., Halleck, P.M., Mathews, J.P., 2011. Inducing fractures and increasing cleat apertures in a bituminous coal under isotropic stress via application of microwave energy. Int. J. Coal Geol. 88, 75–82.
11